多个 ifelse 语句 - 有些不起作用答案

【问题标题】：multiple ifelse statements - some not working多个 ifelse 语句 - 有些不起作用
【发布时间】：2021-08-05 02:08:27
【问题描述】：

对 R 非常陌生，我正在尝试根据不同的条件创建一个变量“准确度”：

如果第一行是 Image2，并且从第 2 行到第 40 行中选择了 image2，则分配 1。如果第一行是 Image2，并且从第 42 到 80 行中选择了 image1，则分配 1。如果第一行是 Image1，并且 image1 是从第 2 行到第 40 行中选择然后分配 1。如果第一行是 Image1，并且从第 42 到第 80 行中选择 image2 然后分配 1

如果都不满足，则赋值为 0。

前 2 条语句可以正常工作，但后 2 条语句不行，任何建议都将不胜感激

data1$Accuracy <- ifelse(data1$Correct_image[1] == "Image2" & between(data1$Trial_number, 2, 40) & data1$Image_selected == "image2", 1,
                  ifelse(data1$Correct_image[1] == "Image2" & between(data1$Trial_number, 42, 80) & data1$Image_selected == "image1", 1,
                  ifelse(data1$Correct_image[1] == "Image1" & between(data1$Trial_number, 2, 40) & data1$Image_selected == "image1", 1,
                  ifelse(data1$Correct_image[1] == "Image1" & between(data1$Trial_number, 42, 80) & data1$Image_selected == "image2", 1, 0))))

【问题讨论】：

一种可能性是使用case_when，它允许您向量化多个 if else 语句。
如果您创建一个小的可重现示例以及预期的输出，这将更容易提供帮助。阅读how to give a reproducible example。

标签： r if-statement variables

【解决方案1】：

所有条件都计算为 1 或 0，因此您也可以将它们全部放在一个（虽然很长）ifelse 调用中：


data1$Accuracy <- ifelse((data1$Correct_image[1] == "Image2" & between(data1$Trial_number, 2, 40) & data1$Image_selected == "image2")|
                                 (data1$Correct_image[1] == "Image2" & between(data1$Trial_number, 42, 80) & data1$Image_selected == "image1")|
                                 (data1$Correct_image[1] == "Image1" & between(data1$Trial_number, 2, 40) & data1$Image_selected == "image1")|
                                 (data1$Correct_image[1] == "Image1" & between(data1$Trial_number, 42, 80) & data1$Image_selected == "image2"),1,0)

没有看到数据就不可能走得更远。

例如，如果 Correct_image[1] 是一个 char 变量，而 Correct_image[1] 和 Image_selected 只有两个级别，则可能有一些更简单的解决方案，例如：

#1 - a little modification so 'image1' and 'Image1' are no longer different:

data1$Correct_image[1]<-tolower(data1$Correct_image[1])
#2 simpler conditions in case_when():

data1$Accuracy<-case_when(data1$Correct_image[1]==data1$Image_selected & between(data1$Trial_number, 2, 40) ~ 1,
                          data1$Correct_image[1]!=data1$Image_selected & between(data1$Trial_number, 42, 80) ~1,
                          TRUE ~ 0)

【讨论】：

【解决方案2】：

正如@yang-wu 建议的那样，dplyr 的 case_when 非常适合这种情况。

library(dplyr)
data1 %>%
  mutate(Accuracy = case_when(
    Correct_image == "Image2" & between(Trial_number,  2, 40) & Image_selected == "image2" ~ 1,
    Correct_image == "Image2" & between(Trial_number, 42, 80) & Image_selected == "image1" ~ 1,
    Correct_image == "Image1" & between(Trial_number,  2, 40) & Image_selected == "image1" ~ 1,
    Correct_image == "Image1" & between(Trial_number, 42, 80) & Image_selected == "image2" ~ 1, 
    TRUE ~ 0))))
  ))

【讨论】：

如果这不起作用，如果没有一些代码来生成示例数据和预期输出，将很难解决您的问题。
另外，这里的“一些不工作意味着”具体是什么意思？您的预期是什么？结果与此有何不同？